As a result of, inter alia, the increased use of multimedia computers, the demand for higher density storage in optical media is increasing. The capacity of the optical media (i.e., optical disks and/or the like), which is typically based on the density of the information in the optical media, has substantially increased in recent years and exponential growth in the capacity of optical media is planned over the next few years. As an example of the density increase, the 4X generation of optical media commonly has a capacity of about 2.6 GB and the more recently developed 8X generation commonly has a capacity of about 5.2 GB.
When increasing the capacity of an optical disk, the separation of the spiral tracks (each track is typically comprised of a groove between two lands) commonly formed on the surface of the optical disk is substantially reduced such that the individual track lands are typically less than 1 um apart from each other. Numerous marks (also known as domains), the size of which are typically determined by the length of a binary representation of a data field, are commonly recorded in the grooves between the track lands (see FIG. 1). Due to the decreased distance between adjacent tracks on the high density optical disk, the formation of a mark substantially in a groove between two adjacent track lands often becomes increasingly difficult.
To write a mark within a track or to increase the number of marks on an optical disk, a sufficiently small optical beam is typically required. Shorter wavelength lasers and higher numerical aperture lenses for the reading and writing devices typically determine the beam spot size, and consequently, the size of each mark. Thus, to decrease the size of the optical beam, a high power semi-conductor red laser (685 nm) is most often utilized when writing the magnetic code onto the optical disk. However, the 685 nm laser typically provides the shortest wavelength laser beam currently available in the market. Moreover, the numerical aperture is often restricted to be less than about 0.55 in common implementations. Thus, a further substantial reduction in the size of the optical beam by a shorter wavelength or larger numerical aperture written onto the optical disk presents practical problems.
Because of the limitations in reducing the size of the focused optical beam, the larger beam spot often extends beyond the width of a single groove in a high density optical disk and, at times, senses a signal pattern from the adjacent groove (see FIG. 1), thereby resulting in a problem known as adjacent track crosstalk (ATC). ATC typically becomes a more pronounced problem when writing longer period (T) data onto a high-density optical disk.
More particularly, when writing a long mark (i.e., 5T and larger) onto an optical disk, an increased laser power is often required because of the additional pulses in the pulse sequence. For each pulse, the optical media is usually heated then cooled. However, for subsequent pulses, the optical media is usually heated again before the cooling process for the previous pulse is sufficiently completed. Consequently, the resulting mark is typically wider, as shown in FIG. 2, because of the lack of proper cooling in between pulses. Moreover, the excess heat commonly results in blooming. Blooming is a common problem whereby the excess heat increases the width of the mark, as is evident in FIG. 2. The problems associated with ATC are often expressed when reading adjacent to tracks with excessively wide marks.
Specifically, when reading from a disk, the laser beam commonly analyzes each mark within the track. When ATC exists, the data contained within the excessively long mark is partially sensed by the read focused spot when the reading process occurs on an adjacent track. The amount of crosstalk that is coupled to the read focused spot is typically proportional to the length of the mark and spacing between marks. For example, and as shown in FIG. 1, when reading a 2T pattern, the read focused spot may sense an 8T pattern located in an adjacent track. If the amount of the adjacent track signal pattern which is sensed by the read focused spot is greater than about 10% of the 2T pattern signal amplitude, excessive jitter typically occurs, thereby decreasing the reliability in the read channel. Consequently, because 8T is the longest mark and 2T is the shortest mark, the maximum amount of crosstalk will often enter into the smallest signal amplitude of the 2T pattern. The typical crosstalk measured on a track separation of about 0.85 um and a 2T mark length of about 0.53 um are shown below:
Data Pattern Crosstalk 2T mark, 2T space -36 dB 3T mark, 3T space -33 dB 4T mark, 4T space -30 dB 5T mark, 5T space -26 dB 6T mark, 6T space -22 dB 7T mark, 7T space -18 dB 8T mark, 8T space -14 dB
FIG. 2 shows a typical prior art recording method used in obtaining these crosstalk measurements by showing an exemplary write clock corresponding to exemplary encoded data (each line "A") having respective exemplary write pulses on each line "C" and exemplary written magnetic domains for 2T-8T marks.
An apparatus and method is needed for reducing the width of larger period marks on high density optical disks, thereby substantially reducing blooming and ATC.